Crohn's disease (CD) and ulcerative colitis (UC) affect over 1.4 million Americans. Over the past 15 years, considerable progress has been made in the therapy of CD and UC with monoclonal antibodies against tumor necrosis factor (anti-TNF) representing the most significant therapeutic breakthrough. However, many patients fail to respond adequately. Recent genome wide associations studies (GWAS) have significantly increased our understanding of the pathogenesis of CD and UC. The role of these genetic variants in predicting response to anti-TNF therapy is yet to be defined adequately. In addition, significant safety concerns remain with the use of these agents with serious infections occurring in up to 10% of patients. There is a strong genetic component to susceptibility to infections, in particular involving polymorphisms in innate immunity. Given the key role of innate immunity in IBD pathogenesis, whether such shared polymorphisms also influence risk of infections on anti-TNF therapy has not been examined previously. Thus, the development of tools to define likelihood of treatment response and infections are a key unmet need in the field. Our overarching objective is to identify clinical and genetic predictors of both short-term and long-term response as well as infectious complications of anti-TNF therapy. To achieve these aims, we will utilize the complementary strengths of two large cohorts - a prospective registry cohort of over 1200 patients with CD or UC, and a novel EMR-based cohort of over 11,000 patients with CD or UC linked to a biospecimen repository. This proposal utilizes the mentorship and significant expertise in key areas including analytical genetics, translational immunology, and bioinformatics and allows the candidate to develop fundamental skills in these novel content areas and research methods. The access to expertise at the NIH-funded center Informatics for Integrating Biology and the Bedside, the Center for Study of Inflammatory Bowel Disease, and the Broad Institute is a key strength of this application. This career development award will be critical in providing training in bioinformatics and genetic epidemiology, thus contributing to the candidate's long-term goal of establishing an independent IBD research career with focus on defining at-risk cohorts and personalizing therapy. The training component of the award includes graduate-level courses in the relevant content areas in genetic epidemiology, and bioinformatics at the Harvard Medical School, School of Public Health, and Massachusetts Institute of Technology. The candidate will continue to actively engage with an established group of multidisciplinary researchers who have demonstrated the utility of the EMR-based approach to accurately define autoimmune diseases, link it to biospecimen collection, and use it successfully for genotyping-phenotype research. This will help foster collaborative development of future proposals as the work proposed in this application has the potential for applicability in other cohorts.